Outline of case study presentation

a)Introduction

b)Research Question

c) Research hypothesis

d)Normality test

e) Statistic hypothesis (Ho and Ha)

f) Descriptive analysis(bar chart/ pie chart /mean,

standard deviation, median,IQR)

g) Inferential analysis (t, F, r, X2)

h)Conclusion (follow APA style)

The Pearson Correlation (r)

Is an index of relationship between two variables

Is represented by the symbol ‘r’

Reflects the degree of linear relationship between two

variables

It is symmetric. The correlation between x and y is the

same as the correlation between y and x.

It ranges from +1 to -1.

Significance of the Test

Correlation is a useful technique for investigating the

relationship between two quantitative, continuous

variables. Pearson's correlation coefficient (r) is a

measure of the strength of the association between the

two variables.

The Pearson Correlation (r)

A perfect

linear

relationship,

r = 1

A perfect

negative

linear

relationship,

r = - 1

A correlation

of 0 means

there is no

linear

relationship

between the

two variables,

r = 0

A correlation of .8 or .9 is regarded as a high correlation

there is a very close relationship between scores on

one of the variables with the scores on the other.

A correlation of .2 or .3 is regarded as low correlation

there is some relationship between the two variables,

but it’s a weak one.

Formula Used (for manual calculation):

Where:

x : Independent variable (BMI)

y : Dependent variable (Systolic)

r = Ʃxy

(Ʃx2) (Ʃy2)

Statement of the Problem

To correlate the relationship between the following demographicvariables. For our group, we have to find the correlation betweensystolic and the BMI (weight and height) among the adults . Does thelevel of systolic being influence by the BMI level of a person?

a. Ageb. Racec. Smoking habitd. Level of COe. Systolicf. Diastolicg. Waisth. HipI . Highj . Weightk. Glucosel. Cholesterol

BMI = Weight (kg)Height 2(m)

Is there a correlation between the systolic level and the BMI

reading of 96 adults?

BMI = Independent VariableSYSTOLIC = Dependent Variable

RESEARCH HYPOTHESIS

There is a linear relationship between the systolic level and the BMI reading.

Ho : Data is normally distributed

Ha : Data is not normally distributed

For normality test, we referred to Shapiro –Wilk (sample size is < 100). From

the Normality table, W (Sig. Shapiro-Wilk) = 0.099 for BMI , 0.281 for Systolic.

Since Shapiro-Wilk sig value is more than alpha value (α > 0.05) thus the

normality assumption is not violated.

Conclusion : Data is normally distributed.

PROBLEM:

Is there a correlation between the

systolic level and the BMI reading of

96 adult?

n = 96

Statistic Hypothesis (Ho and Ha)

Ho: = 0

Ha: ≠ 0

: Rho

Scatter Plot

This scatterplot shows that there is a linear relationship between the two variables.

Putting the Formula together:

ANALYSIS IS DONE USING SPSS VERSION22

At alpha a = 0.05 compare the sig. value from correlation table in

SPSS output.

The Pearson Correlation (r) between Systolic and BMI is 0.228. This

correlation is statistically significant (Sig < 0.05)

Pearson Correlation = 0.228

N= 96

Sig (2-tailed) = 0.025

r = 0.228 , The systolic and BMI are moderately correlated

Coefficient of correlation, r2 = 0.052x100% = 5.2%

That is 5.2% of the variability in systolic can be predicted

by variability of BMI

Effect Sizer > 0.5 strong correlation

0.2< r < 0.3 moderate

correlation

0< r <0.1 weak correlation

Conclusion

To assess the size and direction of the linear relationship

between systolic and BMI, a bivariate Pearson’s product-movement

correlation coefficient ( r ) was calculated. The bivariate correlation

between these two variables was positive but moderate.

r(94) = 0.228 , α , 0.05

Degree of freedom for bivariate correlation are

defined as N – 2 (where N is the total sample)

Prior to calculating r, the assumptions of normality was assessed and

found to be supported. A visual inspection of the box plot and histogram

for each variable confirmed that both were normally distributed. In

conclusion, the Pearson’s Correlation (r) between systolic and BMI is

0.228. This correlation is moderately correlated and statistically significant

(Sig. <0.05).

Rerefences

Allen, P., & Bennett, K. (2010). PASW STATISTIC BY

SPSS: A PRACTICAL GUIDE VERSION 18.0. Australia:

CENCAGE Learning.

Cheong, W. K., & Kai Lit, P. (2006). Statistics Made

Simple for Healthcare and Social Science

Professionals and Students. Malaysia: Universiti

Putra Malaysia Press.

SUGGESTION

Although it shows that there is a correlationbetween systolic level and BMI with moderatecorrelation between this two variables (r= 0.228),but the assumption of linearity cannot be met.

When assumption of linearity cannot be met, thesuitable alternative to the Pearson’s correlation isSpearman’s Rho and Kendall’s Tau- B.